A method for super-resolution of face images based on transformer

Abstract

The present invention provides a Transformer-based end-to-end face super-resolution method, including S1: data preprocessing, obtaining image block sequences; S2: using convolutional neural network as an encoder to extract image local features; S3: Transformer-based The encoder module uses the self-attention mechanism to extract global remote features based on the sequence of image blocks; S4: Combines global and local features to implement an end-to-end face super-resolution method. Through the technical solution of the present invention, the content of the present invention mainly includes two parts, one is to process two-dimensional images, using the self-attention mechanism to extract the non-local long-range dependent information of the image sequence; the other is to use the local features extracted by the convolution operation at the same time , the two are combined as the input of the super-resolution decoder, the purpose is to reduce the complexity of model training by using the end-to-end learning method while enhancing the image features.

Description

technical field [0001] The present invention relates to the technical field of image processing and human face super-resolution, in particular, to a Transformer-based method for super-resolution of human face images. Background technique [0002] Face super-resolution (face illusion) is a super-resolution problem in a specific field. Specifically, the detailed information of the input low-resolution face is enhanced through super-resolution technology, and then the corresponding high-resolution face is inferred or restored. image. As we all know, the face is a biological feature, and its related applications are widely used in the current society, such as face recognition systems, criminal investigation, entertainment and other fields, but are limited by the actual application scenarios obtained or generated The quality of face images is uneven, and some image noise will inevitably be superimposed, which makes the quality of face images poor. At the same time, face super-re...

AI Technical Summary

Problems solved by technology

For example, CNN's convolutional layer can effectively extract local image features such as skin color, eye size, and nose shape, but remote related features such as "the nose is above the mouth" and "the distance between the eyebrows and the eyes" cannot. Use multiple convolution kernels to capture

Images